Error in function fmq6to5ms

[dave452]

In function fmq6to5ms, which is used for DecMqUpMS there is a nl (number of loops) in the expression when I think it should be nlq (number of light flavours).
Line 3578 in CRunDec.3.1.cpp.
+nl*(-1327./11664. ... -> +nlq*(-1327./11664. ...

[peterstangl]

I think you're right. But I guess this should be addressed by the original authors of CRunDec (https://www.ttp.kit.edu/preprints/2017/ttp17-011). I think CRunDec is not directly available on GitHub, but at least one of the authors, @Herren, might be active here.

Btw, I've seen that there is a new version CRunDec 3.1.1 from September 2025, which is not yet available through rundec-python (I have created issue #22 as a reminder). But this new version still has the same issue.

[dave452]

This issue has happened before for a different function in CRunDec, from one of the comments :
Dec 2015: bugfix: nl -> nlq in fas6to5ms, ported DecAsDownSI/DecAsUpSI from RunDec

[Herren]

Indeed, that seems to have gone wrong when converting things. I'll fix things in the next days (and will get rid of nl in favour of nloops) and have the TTP folks update the tarball and include a v3.1.2. Will let you know once that is done.

Thanks for pointing it out!

[DavidMStraub]

and have the TTP folks update the tarball and include a v3.1.2

Pleeeeease Github repo for RunDec instead 😺

[Herren]

Will do!

[peterstangl]

@Herren, as a suggestion for the Github repo, you could make an initial commit using CRunDec3 including a Github release of version 3.0. And then make additional commits and releases for versions 3.1, 3.1.1, and 3.1.2. Like this, one could easily switch between versions using the release tags, and one would directly see the changes in the diffs of the commits.